AT&T Earnings Highlight a Bigger Bet on Agentic AI

AT&T used its second-quarter 2026 earnings call to outline a long-term strategy centered on agentic AI, arguing that autonomous software systems will fundamentally reshape enterprise network demand and eventually justify premium pricing for guaranteed wireless connectivity. The strategy comes alongside strong quarterly results. Revenue rose 2.3% year over year to $31.6 billion, while adjusted EBITDA increased 5.2% and adjusted earnings per share climbed to $0.65 from $0.54 a year earlier. CEO John Stankey used the earnings call to position agentic AI as a structural shift rather than a passing technology trend. Traffic Patterns Are Changing Traditional mobile networks were built primarily to deliver downstream traffic such as video streaming and web browsing. Agentic AI systems reverse that pattern by continuously transmitting sensor data, telemetry, and inference requests back to cloud infrastructure. Stankey said AI agents already generate substantially more network traffic than humans performing comparable tasks, requiring carriers to rethink how wireless networks are designed. "Today, industry research shows AI agents generate up to 450% more total traffic per task than a human performing the same work," Stankey said during the earnings call. He added that agentic AI could generate nearly nine times more enterprise network traffic and almost seven times more consumer traffic by 2035. Stankey cited drones, autonomous vehicles, robotics, and augmented reality glasses as examples of applications that require low-latency, uplink-heavy connectivity. He also linked that shift to AT&T's spectrum portfolio, arguing that the company's low-band spectrum enables reliable upstream performance by reaching deeper inside buildings while maintaining consistent connectivity. Capital Spending Reflects the Strategy The expected shift in network demand is influencing AT&T's investment priorities. The company invested $6.1 billion during the quarter, up from $5.1 billion a year earlier, and reaffirmed plans to spend between $23 billion and $24 billion in 2026. It also expects to add 8 million new fiber locations this year. Separately, AT&T is completing its acquisition of additional wireless spectrum licenses from EchoStar, which Stankey said will strengthen its spectrum portfolio. CFO Pascal Desroches said revenue from AT&T's Advanced Connectivity segment, which includes fiber and wireless services, increased 5.1% year over year, while EBITDA grew 8%. He attributed the gains to continued revenue growth and ongoing cost-transformation efforts. A New Enterprise Pricing Model AT&T also suggested that agentic AI could reshape how enterprise wireless services are sold. Drawing a comparison with managed VPN services, Stankey said businesses deploying autonomous systems will increasingly require guaranteed wireless performance. "The dynamics of VPN and why those engineered and highly managed networks were important for a period of time in the wireline space will start to become more prevalent in the wireless space," he said. The comparison suggests AT&T sees AI driving demand not only for additional bandwidth but also for premium connectivity services backed by performance guarantees. Industry Signals Point the Same Way AT&T's outlook aligns with broader industry research. A 2026 Cisco and Foundry survey of more than 3,400 IT and networking leaders found AI-driven network traffic is expected to triple within three years. Ericsson's June 2026 Mobility Report found that 43 of the 55 service providers it surveyed are already seeing uplink traffic grow faster than downstream traffic, reversing a pattern that has defined mobile networks for decades. At Huawei's June 2026 Mobile Broadband Forum, Deputy Chairman David Wang said, "The mobile communications industry is entering the age of intelligence." "We can use AI to upgrade networks across the board. And with AI-native architecture for next-generation core networks, we can more effectively support agentic services," he added. Gartner, meanwhile, forecasts that about 70% of enterprises will adopt agentic AI within IT operations by 2029. However, it also expects more than 40% of agentic AI projects to be canceled by 2027 because of cost pressures, governance issues, and limited business value. Those projections also underscore the uncertainty surrounding AT&T's thesis. While autonomous AI systems could eventually require significantly greater network capacity, enterprise deployments remain limited by implementation costs and operational complexity. How quickly organizations move from pilots to production will determine whether telecom providers see the surge in traffic they are preparing for. Not everyone believes substantial new network investment is immediately necessary. Light Reading has argued that U.S. carriers already have sufficient capacity for most foreseeable AI workloads and that AT&T's elevated capital spending is driven primarily by

AT&T Earnings Highlight a Bigger Bet on Agentic AI

Traffic Patterns Are Changing

Keep reading. It's free.

Related stories